Cigarette smoking is the leading cause of preventable death in the United States. Smoking produces over 440,000 deaths each year in this country and generates an estimated $167 billion in annual health-related economic losses. Available methods of smoking assessment (e.g., self-report, portable puff-topography instruments) do not permit the collection of accurate, non-reactive measures of smoking behavior that capture real-time smoking frequency and comprehensive within-cigarette puff topography. The objective of this project is to develop a non-invasive wearable system (Personal Automatic Cigarette Tracker - PACT) that is completely transparent to the end user and does not require any conscience effort to achieve reliable monitoring of smoking behavior in free living individuals. Methodologically, PACT will consist of two major components: 1. Wearable sensors. Miniature sensors integrated into the clothing will monitors the breathing and activity patterns of individuals. The signals from the sensors will be processed and recognized to identify and objectively characterize each individual puff. 2. Software for signal processing and pattern recognition. Automatic computer software will analyze sensor signals and detect patterns uniquely identifying smoking events. Objective metrics such as number of puffs and inter-puff interval will be extracted. The software will be based on the state-of-art machine learning methods. The development of the PACT system will be addressed in four specific aims: Specific Aim 1: Develop a wearable sensor system comprised of a breathing sensor integrated into conventional underwear and a hand gesture sensor integrated into a hand bracelet. Specific Aim 2: Collect sensor data from individuals wearing the instrumented system and performing everyday activities (including smoking) in laboratory conditions. Specific Aim 3: Develop pattern recognition methods to recognize individual puffs and smoke inhalation. Specific Aim 4: Evaluate the utility and sensitivity of the wearable sensor PACT system and pattern recognition method in people smoking in the natural environment. This set of Specific Aims will validate lead to creation of a unique wearable device capable of objective characterization of smoking behavior. PUBLIC HEALTH RELEVANCE: Cigarette smoking is the leading cause of preventable death in the United States. Smoking produces over 440,000 deaths each year in this country and generates an estimated $167 billion in annual health-related economic losses. The goal of this research is to develop a non-invasive wearable system (Personal Automatic Cigarette Tracker - PACT) that is completely transparent to the end user and does not require any conscious effort to achieve reliable monitoring of smoking behavior in free living individuals. The PACT device will provide an accurate and precise measure of real-world smoking. The device can provide the user and health professional feedback on the frequency of smoking and inhalation patterns (such as depth of inhalation and smoke holding) throughout the day in their home and community. This information can be used to inform behavioral strategies in smoking cessation programs. The data collected by PACT can also provide an objective method of assessing the effectiveness of behavioral and pharmacological smoking interventions.